Video analysis is powered by a large language model (LLM) and other AI methods that perform intelligent video inspection and reporting.

Key capabilities:

Screenshot extraction: Identify and capture key frames from the video.
Issue Detection: Analyse key frames for potential visual discrepancies e.g. missing or distorted textures or mixed language to detect potential problems.
Report Generation: Generate bug reports that include screenshot, timestamp and issue descriptions.

The LLM uses a curated dataset of bug reports to ensure that outputs are consistent. Over time, its accuracy improves as more feedback and issue data are incorporated into the system to ensure the issue detection is customized to the individual project.

To learn how to provide feedback on analysis results, see Give feedback with example reports.

Previous	Next
Video analysis	Automatic analysis of run videos